The product composition of bilateral trade encapsulates complex relationships about comparative advantage, global production networks, and domestic politics. Yet, despite the availability of product-level trade data, most researchers rely on either the total volume of trade or certain sets of aggregated products. We develop a new dynamic clustering method to effectively summarize this massive amount of product- level data. The proposed method classifies a set of dyads into several clusters based on their similarities in trade profile — the product composition of imports and exports — and captures the evolution of the resulting clusters over time. We apply this method to two billion observations of product-level annual trade flows, encompassing over 600 products for more than 59,000 directed dyads from 1962 to 2014. We show that typical dyadic trade relationships evolve from zero trade to inter-industry trade and then to intra- industry trade. Finally, we illustrate the critical roles of our trade profile measure in the analysis of trade competition and bilateral investment treaties as well as the investigation of trade and security linkages.

Prof. Steven Liao